Group Parametric Independent Colored Sources: Detection of Hidden Brain Activities from Groups of High-Dimensional Neuroimaging Data

• Activity Number: 481
• Type: Topic Contributed
• Date/Time: Wednesday, August 12, 2015 : 8:30 AM to 10:20 AM
• Sponsor: Section on Statistics in Imaging
• Author(s): Dong Wang* and Seonjoo Lee and Haipeng Shen and Young Truong
• Companies: The University of North Carolina at Chapel Hill and Columbia University and The University of North Carolina at Chapel Hill and The University of North Carolina at Chapel Hill
• Keywords: Group independent component analysis ; Neuroimaging data ; Spectral analysis ; Whittle likelihood

Abstract

Independent component analysis (ICA) is a popular powerful method for detecting hidden brain activities from neuroimaging data. Motivated by analysis of groups of such high-dimensional imaging data, we develop a group ICA framework in the frequency domain through Whittle log-likelihood maximization. Our method starts with efficient population value decomposition that makes our approach scalable to massive neuroimaging data, and then models each temporally-dependent source signal via parametric linear processes. The superior performance of our approach is demonstrated through simulation studies and the ADHD200 data.